CADER: Confidence-Aware Dynamic Evidence Reasoning for Long-Video Understanding
2026-07-27 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors present CADER, a method that helps computer systems understand long videos more efficiently. Instead of always using complicated tools for every question, CADER first tries a simple approach and checks its confidence in the answer. If it's confident, it stops early; if not, it uses more detailed steps to find better evidence in the video. This way, easy questions get quick answers, and harder ones get more attention, improving accuracy and speed. Their tests show CADER works well compared to other systems, even without extra training.
Long-video understandingVision-language modelsVideoQAConfidence estimationLogit-marginTemporal evidenceTool-augmented reasoningAdaptive inferenceChain-of-thought supervisionSemantic verification
Authors
Jinlong Yang, Wenhao Zhang, Kuanwei Lin, Sijie Cheng
Abstract
Long-video understanding increasingly relies on large vision-language models and tool-augmented reasoning, but most systems apply the same inference procedure to every example regardless of difficulty. This uniform strategy invokes unnecessary tool-assisted processing for easy questions and provides limited control when difficult questions require fine-grained temporal evidence. We propose CADER (Confidence-Aware Dynamic Evidence Reasoning), a training-free framework for adaptive and reliable long-video reasoning. CADER first performs global reasoning over uniformly sampled frames and estimates answer confidence with a logit-margin signal, allowing high-confidence examples to exit early. For uncertain examples, CADER activates a second-stage tool-augmented loop that combines temporal cropping, lightweight semantic verification, and Relevance-Guided Resampling to progressively localize question-relevant evidence. This design treats tool use as a sample-level decision: a single global pass handles easy cases, while additional reasoning is reserved for examples where uncertainty suggests that more evidence is needed. Experiments on multiple VideoQA benchmarks show that CADER improves long-video reasoning while bypassing Stage~2 for high-confidence samples. Moreover, when applied to a backbone trained only with tool-free chain-of-thought supervision, CADER achieves competitive performance against specialized tool-augmented frameworks, suggesting a practical inference-time route for adaptive long-video reasoning.